77

Bloom labAugust 20, 2025 at 1:22 AM EDT

In new study led by @bdadonaite.bsky.social, we measure how spike mutations affect function & antigenicity of spike of KP.3.1.1 strain of SARS-CoV-2. Sheds light on how key neutralizing epitopes are changing & importance of RBD up/down motion. biorxiv.org

We examined spike of KP.3.1.1, a strain from late 2024 / early 2025 similar to current variants. KP.3.1.1 & other recent variants have >60 spike amino-acid mutations relative to early pandemic strains, as spike has evolved at extraordinary rate of >10 mutations/year on average.

We previously developed pseudovirus deep mutational scanning (pubmed.ncbi.nlm.nih.gov), which uses non-replicative viral particles to safely study spike mutations. Here we used approach to measure how mutations to KP.3.1.1 spike affect five phenotypes, as shown below.

We first measured how all mutations affect cell entry; see this interactive page for those data: dms-vep.org A number of mutations that have spread recently are more tolerated in KP.3.1.1 than older XBB.1.5 variant, suggesting epistatic shifts favored their emergence.

We next measured how mutations affect full-spike ACE2 binding. Most mutations had similar impacts in KP.3.1.1 and the older XBB.1.5 spike, but some recent mutations (eg, A435S and Q493E) enhance ACE2 binding KP.3.1.1 after impairing it in XBB.1.5.

We measured how mutations affected neutralization by sera collected from humans before or after vaccination or infection w recent JN.1-descendant variant. Key sites of escape are shown below; some of the sites (eg, 475 and 478) are mutated in very recent variants.

Despite imprinting, in some individuals recent infection or vaccination appreciably shift immunodominant neutralizing epitopes. So new exposures are altering neutralizing serum antibody repertoire, although our data do not define mechanism (see preprint for hypotheses).

Some mutations outside the RBD have a strong effect on serum antibody neutralization. But for ACE2-distal or non-RBD mutations, there is a strong tradeoff between serum antibody escape and ACE2 binding as shown below.

This tradeoff between serum antibody escape and ACE2 binding is because mutations that put the RBD more up enable ACE2 binding but also promote antibody binding. Mutations that put the RBD more down do the opposite.

We used this fact to estimate how much mutations at each site affect RBD up-down motion, as shown below. Many of these sites have mutated during SARS-CoV-2 evolution in humans, demonstrating importance of RBD motion and its effects on ACE2 binding & antibody neutralization.

Finally, we measured how mutations affect neutralization by three relevant monoclonal antibodies. As shown below, all antibodies adversely affected by mutating site 505 which fortunately remains highly constrained for ACE2 binding. We discuss this interesting site more in preprint.

All data available in interactive form at dms-vep.org and we encourage exploration of that site. Study led by the incomparable @bdadonaite.bsky.social, w help from Sheri Harari, Brendan Larsen, Lucas Kampman, Alex Harteloo, Anna Elias-Warren, & Helen Chu.

The final published version of this study is now available in Journal of Virology at journals.asm.org

1 / 10